Preparing Future Designers for Human-AI Collaboration in Persona Creation
Toshali Goel, Orit Shaer, Catherine Grevet Delcourt, Quan Gu, Angelora Cooper
Abstract
This paper presents findings from an exploratory study investigating the use of AI text-generation tools to support novice designers in persona creation. We conducted a workshop with 22 undergraduate students enrolled in an introductory human-computer interaction course, who were instructed to use GPT-3 in the creation of personas. These novice designers were able to use GPT-3 to iterate to produce satisfactory personas, particularly when providing detailed prompts. Our findings suggest that personas created with GPT-3 assistance were mostly comparable to those created manually but rated lower on some evaluation dimensions. The study also reveals merits and concerns of using GPT-3 for persona creation. Based on our findings, we propose recommendations for novice designers on how to use text-generative AIs to create personas effectively and responsibly.